Research On Intelligent Mobile Spray Robotic System For Field Crops

With the development of agricultural mechanization and intelligent robot, modern agricultural machinery has gradually turned to intelligence agriculture, facility agriculture and high-efficiency agriculture. High-efficiency, intelligence and accuracy are becoming the developmental goal of modern agricultural machinery. This discipline combines machinery, electronics, control theory, plant protection and the most cutting-edge research of many other fields. Small mobile fogging machine, to a certain extent, fills the pest control of tall crops of corn, sorghum and sugar cane which is difficult to control in the periods of middle and late and has great research space and research significance.Pest control is one of the core problems of modern agricultural research. In order to achieve the small mobile fogging machine sprays uniform pesticides in the field automatically. The core technologies of a mobile robot's walking in crop inter-row independently and uniform spray of large area were researched on the corns of later period which were planted by agricultural machinery. The system of small fogging machine in crop inter-row was researched and developed independently. The method of monocular visual path planning was proposed which was based on the radial basis function. The spray droplets sedimentary characteristics of many lines were carried on experiment and analyzed in different speeds.The main research contents and results are as follows:1) The small mobile robot with high obstacle capability and the system of the droplets atomization were studied. The obstacle ability of the small vehicle was greatly improved by the optimization of the chassis structure. The steering mechanism working in narrow environment of crop inter-row was designed by the optimization of steering trapezoid. At the same time, uniform spray pattern was obtained by the research of the thermal spray system of fogging machine and the study of nozzle in spray line.Finally, a fogging machine prototype was developed which met the need of motion of the wild field crop inter-row.2) Unmanned, compact agricultural machines, therefore, are ideal for such field work. This paper describes a method of monocular visual recognition to navigate small vehicles between narrow crop rows. Edge detection and noise elimination were used for image segmentation to extract the stalks in the image. The stalk coordinates define passable boundaries, and a simplified RBF-based algorithm was adapted for path planning to improve the fault tolerance of stalk coordinate extraction. The average image processing time, including network latency, is 220ms. The average time consumption for path planning is 30ms. The fast processing ensures the top speed of 1.2m/s of our prototype vehicle. When operating at the normal speed (0.7m/s), the rate of collision with stalks is under 6.4%. Additional simulations and field tests further proved the feasibility and fault tolerance of our method.3) Fogging machine equipped on the vehicle has the ability to move quickly in the field of spray. In addition, it has a large number of labor-saving; improve spraying efficiency and other advantages. Droplet distribution uniformity, coverage rate and effective distance are important indicators of this fogging machine. A method of semi-quantitative study of the distribution of droplet deposition in the field by water-sensitive paper is presented. The reaction region of water-sensitive paper distinguished from paper background by irradiation of 450-455nm wavelength circular pattern light, which improve results about binary image segmentation. Confidence interval estimation method for removing noise in the collection datao Data analyses for upper and lower limits of confidence interval reduce systematic errors and improve detection accuracy. Repeat the measurement error of this approach for the droplet coverage calculation is less than 5%; Repeat the measurement error is also less than 5%for droplet size measurement when the droplet coverage rate is 10-20%. The fogging machine droplet distribution uniformity, coverage and effective distance are analyzed at a speed of 0.6,0.9,1.1,1.3,1.5 m/s situation, respectively, by using this method. When the speed under 0.6-1.2m/s, the effective distance are about 6-8m (droplet coverage rate more than 10%). When the speed is increased to 1.5m/s, the effective distance is only 5m. Moreover, fog machine droplet coverage rate gradually decreasing with increasing distance, droplet coverage rate divergence increases with increasing distance.Finally, tests of droplets deposition of many lines were taken using the experimental prototype in Guoyang and Huaiyuan. The uniformity and coverage of droplets deposition were tested in different speeds. The precision applying pesticide was optimized and the spray efficiency of prototype was analyzed. The results provided experimental basis for the further implementation of precision applying pesticide of small mobile fogging machine in the field and an experimental reference for improving the efficiency of pesticide spraying.